|
|
||||||||
Ann Thorac Surg 2002;74:13-18
© 2002 The Society of Thoracic Surgeons
a Department of Cardiothoracic Surgery, Stanford University School of Medicine, Stanford, California, USA
* Address reprint requests to Dr Vricella, Department of Cardiothoracic Surgery, Stanford University School of Medicine, Falk CVRB, Stanford, CA 94305-5407, USA
e-mail: vricella{at}stanford.edu
Presented at the Forty-eighth Annual Meeting of the Southern Thoracic Surgical Association, San Antonio, TX, Nov 810, 2001.
| Abstract |
|---|
|
|
|---|
Methods. Sixty-four patients with cystic fibrosis underwent heart-lung transplantation (n = 22, 34.4%) or BLTx (n = 42, 65.6%) between 1988 and 2000. Mean age and weight at transplantation were 29 ± 8 years and 51 ± 11 kg, respectively. Mean follow-up for survivors was 4.4 ± 3.6 years. Immunosuppression regimen included cyclosporine, tapered corticosteroids, azathioprine, and induction therapy with OKT3 (murine monoclonal antibodies) or rabbit antithymocyte globulin. Cardiopulmonary bypass was used in all but 5 patients (7.8%). However, in 8 (19%) of the 42 patients having BLTx, only the grafting of the second lung was performed with cardiopulmonary bypass.
Results. The operative mortality rate was 1.6%. The actuarial survival rates at 1 year, 3 years, 5 years and 10 years were 93.2%, 77.7%, 61.8%, and 48.1%, respectively, with no significant difference between BLTx and heart-lung transplantation. The major hospital complications were pneumonia (n = 11, 17.2%) and bleeding (n = 8, 12.5%). Clinically significant reperfusion injury was observed in 6 patients, 3 of whom required reintubation. Freedom from acute lung rejection beyond 1 year was 47.7%. One patient underwent late retransplantation, and 4 required bronchial stenting. Obliterative bronchiolitis accounted for eight (50.0%) of 16 late deaths.
Conclusions. Though postoperative bleeding and pneumonia are still of concern, satisfactory early and intermediate-term results can be expected in patients undergoing BLTx or heart-lung transplantation for cystic fibrosis. Cardiopulmonary bypass can be used for BLTx with no adverse impact on intermediate and long-term outcomes.
| Introduction |
|---|
|
|
|---|
We present our experience with HLTx and BLTx for treatment of cystic fibrosis over a 13-year period with an operative strategy that preferentially used cardiopulmonary bypass (CPB) for BLTx.
| Material and methods |
|---|
|
|
|---|
Standard immunosuppression with cyclosporine and azathioprine was used in all patients. Initial perioperative methylprednisolone administration was followed (after 1 to 2 weeks) with prednisone tapered to a target daily dose of 0.2 mg/kg. Induction therapy consisted of OKT3 (murine monoclonal antibodies) (before 1993) in 10 patients (15.6%) or rabbit antithymocyte globulin in 52 (81.3%). Two patients did not receive cytolytic therapy and, in 3, rabbit antithymocyte globulin was changed to OKT3 because of patient intolerance.
In case of discordant cytomegalovirus reactivity between donor and recipient (n = 44, 68.8%), a 34-day prophylaxis protocol was used. The regimen consisted of intravenous ganciclovir sodium as well as (after 1996) cytomegalovirus immune globulin (CytoGam); the latter was administered within 72 hours after transplantation and again on postoperative week 2, 4, 6, 8, 12, and 16. Since 1995, the 34-day cytomegalovirus prophylaxis protocol has been supplemented with oral ganciclovir for an additional 6 weeks.
All patients underwent preoperative bilateral maxillary antrostomies and aggressive maxillary flushing with diluted tobramycin sulfate solution, continued monthly in the postoperative period. Antibiotic coverage was directed specifically to the most recent preoperative cultures and adjusted thereafter according to postoperative bronchoscopic findings and sinus microbial isolates. Patients colonized or infected with resistant microorganisms were placed on temporary hold.
Donor and recipient inclusion and exclusion criteria and operative technique have been previously described [2].
Statistical methods were used to make comparisons between the four groups representing different methods of perfusion: HLTx, off-pump BLTx, BLTx with CPB used for grafting the second lung only, and BLTx on CPB during transplantation of both lungs. Statistical analyses were carried out using SAS version 8.1, 1999 (SAS Institute, Inc, Cary, NC). Analysis of variance was performed to compare means between the different groups. Survival estimates and freedom from postoperative events were calculated by life-table analysis (Kaplan-Meier) using the lifetest procedure in SAS. The log-rank test was used for examination of significance (
= 0.05). Results are reported as the mean ± the standard deviation unless otherwise specified.
| Results |
|---|
|
|
|---|
Twenty-two patients (34.4%) underwent HLTx and 42 (65.6%), BLTx. In 5 (11.9%) of the 42 patients undergoing BLTx, CPB was not used; in 8 patients (19.0%) only grafting of the second lung was performed with CPB; and in the remaining 29 patients (69.0%), CPB was used from the beginning of the procedure. Of the 59 patients having transplantation with CPB, 31 (52.5%) received aprotinin. Modified ultrafiltration was used in 20 (33.9%) of these 59 patients, and the average ultrafiltrate was 1,785 mL (range, 9004,000). Heparin-coated CPB circuits and leukocyte-depletion filters each were used in 9 (15.3%) of these 59 patients.
Actuarial survival of all patients is illustrated in Figure 1; the survival rates were 93.2%, 77.7%, 61.8%, and 48.1% at 1 year, 3 years, 5 years, and 10 years, respectively. The operative (30-day) mortality rate was 1.6%, with a total of four hospital deaths (6.3%). One patient died on the second postoperative day of massive pulmonary edema despite extracorporeal circulatory support, and 3 patients died late in the hospital course, 2 of pneumonia and 1 of fungal sepsis and multiorgan system failure. No significant difference in survival was observed between HLTx and BLTx (Fig 2) or between the four different approaches to perfusion (Fig 3).
|
|
|
Fifty-eight (90.6%) of the 64 patients were weaned from mechanical ventilation and discharged. The other 6 patients either required tracheostomy or were intubated at the time of death. Fifty (86.2%) of the 58 patients required a single period of mechanical ventilation, and 8 (13.8%) were reintubated once and eventually weaned without requiring tracheostomy. Mean length of endotracheal intubation was 2 ± 3 days for the 50 patients who needed a single period of mechanical ventilation and 15 ± 8 days for the 8 patients who were reintubated. Two of the 3 patients who underwent tracheostomy were decannulated prior to discharge, and 1 died while on mechanical ventilation. No significant difference in regard to need for reintubation was noted between the four different perfusion strategy groups (p = 0.6).
The vast majority of discharged patients maintained excellent long-term pulmonary function, as indicated by measurements of forced expiratory volume in 1 second and vital capacity (Fig 4). Four (6.7%) of the 60 hospital survivors required bronchial stenting for anastomotic stricture late in the posttransplantation period.
|
|
|
| Comment |
|---|
|
|
|---|
The surgical approach to thoracic transplantation for cystic fibrosis has evolved during the last two decades, as is reflected in this report. Between 1988 and 1992, we performed HLTx in patients with cystic fibrosis because of our experience and confidence with the procedure [6]. To optimize organ allocation, transplantation of the recipients heart was possible for a large number of heart-lung recipients in this series, the so-called domino heart transplantation [7]. As BLTx gained wide acceptance as therapy for primary pulmonary disease [1, 8], we shifted after 1992 toward BLTx as our preferred method of treatment of these patients. Though HLTx remains a well established technique with excellent long-term results [9, 10], BLTx offers several advantages that currently define it as the procedure of choice. Bilateral lung transplantation obviates the necessity to transplant a denervated heart from a brain-dead donor and the possibility of acute cardiac rejection and late graft coronary artery disease, which can be as high as 15% [2]. On the other hand, HTLx may be more expeditious than BLTx and is accomplished without the need of a bilateral thoracosternotomy. Nevertheless, substantial postoperative bleeding associated with the use of CPB and high early mortality after HLTx have been reported [9]. Though acknowledging the size of the cohorts, we have observed no significant differences in actuarial survival curves between HLTx and BLTx (see Fig 2). Furthermore, HLTx often represents the only surgical strategy available in small pediatric patients [11, 12].
Considerable debate surrounds the issue of CPB for BLTx. Several centers [13, 14] have achieved excellent results with very limited use of CPB during BLTx. Others prefer to use CPB routinely if the anticipated extraction and implantation time of the second lung exceeds 1
hours [15]. Those who do not favor the use of CPB point to the detrimental effects of extracorporeal circulation on endothelial permeability and coagulation factors secondary to production of proinflammatory cytokines and humoral mediators [16]. These, in turn, could result in postoperative hemorrhage with increased blood product requirement and reperfusion injury. We recently have used leukocyte-depletion filters at the time of reperfusion to minimize the potential for reperfusion injury. After reperfusion, modified ultrafiltration is routinely performed to reduce pulmonary edema.
It has been our operative strategy to use CPB and single-lumen endotracheal intubation since the beginning of the procedure in all patients seen after 1996. In our opinion, there are several advantages associated with the routine use of CPB. The first is increased exposure of the hilar structures. Patients who have dense adhesions and bronchial collaterals are the ones who most benefit from CPB and optimized exposure rather than manipulation and hemodynamic instability. Second, CPB allows early bilateral pneumonectomies with no hemodynamic or respiratory instability before arrival of the graft and optimal washout of the distal trachea and proximal bronchi without the possibility of contaminating the first implanted lung. The third advantage is avoidance of manipulation; avoidance of the need of the first implanted lung to entirely sustain the circulating volume is intuitively beneficial. Fourth, the ischemic time of the second lung is substantially reduced. In our experience, the average differential ischemic time between the first lung and the second lung was 136 minutes for an off-pump procedure and 69 minutes for an on-pump procedure (both lungs) (p < 0.0001). Lower incidences of cerebrovascular accidents and comparable hemorrhagic complications have been observed compared with other reports [14, 17].
In recent years, we have introduced several modifications to our intraoperative strategy to help minimize the chance of postoperative hemorrhage and the need of reexploration. Aprotinin and heparin-coated circuits are used routinely, and argon-beam coagulation is available. Though we do not have specific data regarding transfusion requirements, we have observed a definite trend toward decreased blood product requirements and lower reexploration rates in recent years.
Mendeloff [13] reported no difference in duration of endotracheal intubation between patients undergoing transplantation with and without the use of CPB in the Washington University experience. This has certainly been our observation as well, with well over 80% of patients requiring a mean duration of endotracheal intubation of only 2 days.
Tracheal or bronchial anastomotic complications (dehiscence or stenosis) appear to occur more frequently in BLTx than HLTx, with a reported incidence ranging between 6% and 26% [13, 14, 18]. We have not routinely telescoped or reinforced airway anastomoses with autologous tissue and have used a simple continuous running 3-0 and 4-0 polypropylene suture for tracheal and bronchial anastomoses, respectively. We have observed no tracheal complications in the 22 HLTx procedures and five stenoses in 84 at-risk BLTx anastomoses (6.0%) (p = not significant). No dehiscences were observed, and all stenoses were treated successfully by dilation or stenting.
Infectious complications are still a major source of concern in the early posttransplantation period; pneumonia develop in many patients. Though apparently counterintuitive, postoperative septic pulmonary and extrapulmonary complications do not appear to occur with higher frequency in patients with cystic fibrosis than in those without this disease [1]. In fact, many centers do not consider colonization with Burkholderia cepacia, multi-drug-resistant pseudomonal strains, or Aspergillus to be an absolute contraindication to transplantation in patients with cystic fibrosis. It has been our practice in the past to perform transplantation only in patients free from resistant infection or colonization. We now routinely obtain periodic synergy testing for panresistant Pseudomonas aeruginosa strains, and patients are kept on the waiting list if sensitivity to two synergistic drugs is documented. Most of these patients are maintained on a regimen of nebulized tobramycin or colistin sulfate. Similarly, patients showing colonization with B cepacia are considered for transplantation only if sensitivity to at least two antibiotics is found on synergy testing. Colonization with Aspergillus is treated with itraconazole, and patients are kept on the waiting list. Aggressive eradication of paranasal and tracheobronchial foci, directed antimicrobial therapy, and aggressive cytomegalovirus prophylaxis are still mainstays of our perioperative antimicrobial strategy.
Future directions of clinical and research efforts on cystic fibrosis include gene therapy [14], xenotransplantation [19], bilateral lower lobe transplantation [14], and use of marginal or nonbeating-heart donors [20].
Currently, pulmonary transplantation represents the sole therapeutic option for patients with end-stage cystic fibrosis, and it can be accomplished with excellent medium-term and satisfactory long-term results. Though postoperative bleeding and pneumonia are still of major concern, an operative strategy that mainly uses CPB does not appear to adversely affect outcome. Our experience confirms that among different diagnostic cohorts, these patients are not at a disadvantage in regard to survival because of the septic nature of their disease. Long-term survivors maintain excellent pulmonary function and quality of life, though prolonged survival is often limited by late infections and bronchiolitis obliterans.
| Acknowledgments |
|---|
|
|
|---|
| Discussion |
|---|
|
|
|---|
DR VRICELLA: In this particular group, 6 patients had evidence of reperfusion injury. However, this needs to be further defined, as patients with cystic fibrosis are usually already infected or have positive cultures. Therefore, the differentiation between reperfusion injury and pneumonia must be made on radiologic grounds, with patchy infiltrates being consistent with the former and lobar consolidation with the latter. Of the 6 patients with reperfusion injury, 3 were reintubated, and 2 required extracorporeal membrane oxygenation. Of the latter, 1 died on postoperative day 2, and 1 was a long-term survivor.
DR FREDERICK L. GROVER (Denver, CO): This was a study with nice results. I would appreciate it if you would talk a little about the use of cardiopulmonary bypass (CPB). At this time, we are trying to get away from it, when we can, in a lot of our coronary work because of the release of cytokines and the proinflammatory reaction. We have taken the opposite approach, and have not utilized CPB for 17% of patients. Survival and complication rates in CPB and non-CPB were similar.
I think that if the operations are done without CPB, it is key to perform bronchoscopy and to wash out the distal tracheobronchial tree before the double-lumen tube is inserted so that you do not run into airway problems. Maybe you could give us your rationale for that.
DR VRICELLA: We have moved to single-lumen endotracheal intubation and institution of CPB from the start in most of our recent recipients of double-lung transplantation. I think that there are several advantages to performing the procedure with CPB. Typically these patients have tremendous vascular adhesions; so exposure is key. Bilateral pneumonectomies can be accomplished with maximal exposure and hemodynamic stability as well as optimal washout of the distal tracheobronchial tree. Other intuitive advantages are the reperfusion of the right lung at low pressure and the considerably shorter ischemic time of the second lung. When reviewing the differential ischemic time between the first and second grafts, we went from a mean of 136 minutes in the five off-pump operations to 69 minutes in those accomplished with the aid of CPB. I also think that we select patients to a worse outcome if we start CPB only when we encounter intraoperative problems. In most series in which CPB is used sporadically, it is done in an emergent situation; specifically, CPB might be started because of bleeding from the pulmonary artery or poor respiratory or hemodynamic performance after reperfusion of the first transplanted lung.
There is certainly a wealth of data regarding release of cytokines and inflammatory mediators with CPB; the routine insertion of leukocyte-depletion filters in the CPB circuit prior to reperfusion as well as other adjuncts previously mentioned is an attempt to address this issue.
DR CLIFFORD H. VAN METER, JR (New Orleans, LA): You have reported excellent results. How old were your patients. Some recent data imply that because of survival on the waiting list versus survival with surgical intervention in pediatric patients, a forced expiratory volume in 1 second of less than 30% of predicted may not be a good indicator for transplantation, and rate of decline might be a better factor. How does your patient population confirm or contrast with that?
DR VRICELLA: The mean age at death of untreated patients with cystic fibrosis is 29 years, the same as that of the recipients in our group. In this series, 5 patients were in the pediatric group, and the youngest received a heart-lung block at 10 years of age.
In regard to your second question concerning organ allocation and impact of transplantation on survival of pediatric patients with cystic fibrosis, recent data from England show a greater than 70% survival advantage for children having transplantation compared with those on the waiting list. I believe children with cystic fibrosis should undergo transplantation if possible.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. S. Ganesh, C. A. Rogers, R. S. Bonser, N. R. Banner, and on behalf of the steering group of the UK Cardioth Outcome of heart-lung and bilateral sequential lung transplantation for cystic fibrosis: a UK national study Eur. Respir. J., June 1, 2005; 25(6): 964 - 969. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |